|
Katharina Prasse I'm a PhD student at the Data and Web Science Group at the University of Mannheim in Germany. I am supervised by Prof. Dr.-Ing. Margret Keuper Prior to this, I have been a researcher at the University of Siegen I have completed my master degree in Data Science at the University of Mannheim. For my Bachelor's degree, I have studied at the University College Roosevelt in the Netherlands. I have completed a liberal arts and science programm with majors in Psychology and Economics, and a minor in Statistics. |
|
ResearchI'm interested in computer vision for real-world applications and explainable AI. I enjoy combining my research with my passion, i.e. climate change and bicycle infrastructure projects. |
ICML 2025 |
DCBM: Data-Efficient Visual Concept Bottleneck Models
Katharina Prasse*, Patrick Knab*, Sascha Marton, Christian Bartelt, Margret Keuper International Conference on Machine Learning (ICML), 2025 paper / code / website
A data-efficient approach for training visual concept bottleneck models that requires fewer labeled concept annotations while maintaining interpretability.
|
WACV 2025 |
I Spy with My Little Eye: A Minimum Cost Multicut Investigation of Dataset Frames
Katharina Prasse, Isaac Bravo, Stefanie Walter, Margret Keuper Winter Conference on Applications of Computer Vision (WACV), 2025 paper / code Applying minimum cost multicut to analyze visual frames in social media datasets, studying how climate change is depicted in online visual content. |
Sci. Comm. 2025 |
Global Dynamics of Climate Change Imagery: Emotional and Engagement Effects Across Visual Frames on Twitter/X
Isaac Bravo, Katharina Prasse, Stefanie Walter, Saffron O'Neill, Margret Keuper Science Communication, 2025 paper Analyzing how different visual frames of climate change imagery on Twitter/X affect emotional responses and user engagement across global audiences. |
Annals ICA 2026 |
Visualizing Climate Change in the Media: A Systematic Literature Review, Challenges, and Future Research
Isaac Bravo, Stefanie Walter, Katharina Prasse, Margret Keuper Annals of the International Communication Association, 2026 paper A systematic literature review of how climate change is visualized across media, identifying challenges and future research directions at the intersection of visual communication and computer vision. |
DAGM GCPR 2023 |
Local Spherical Harmonics Improve Skeleton-Based Hand Action Recognition
Katharina Prasse, Steffen Jung, Yue Zhou, Margret Keuper Pattern Recognition: 45th DAGM German Conference (GCPR 2023), Springer 2024 paper Incorporating local spherical harmonics representations into skeleton-based action recognition improves performance on fine-grained hand gesture classification tasks. |
NeurIPS WS 2023 |
Towards Understanding Climate Change Perceptions: A Social Media Dataset
Katharina Prasse, Steffen Jung, Isaac Bravo, Stefanie Walter, Margret Keuper NeurIPS Workshop on Tackling Climate Change with Machine Learning, 2023 paper A curated social media dataset for studying how people perceive and communicate climate change visually, supporting interdisciplinary research in climate communication and computer vision. |
Conference Presentations |
DCBM: Data-Efficient Visual Concept Bottleneck Models
Katharina Prasse*, Patrick Knab*, Sascha Marton, Christian Bartelt, Margret Keuper ICCV Workshop on Explainable Computer Vision (eXCV), Oct. 2025 DCBM: Data-Efficient Visual Concept Bottleneck Models Katharina Prasse*, Patrick Knab*, Sascha Marton, Christian Bartelt, Margret Keuper DAGM German Conference on Pattern Recognition (GCPR), Sep. 2025, Freiburg im Breisgau, Germany DCBM: Data-Efficient Visual Concept Bottleneck Models Katharina Prasse*, Patrick Knab*, Sascha Marton, Christian Bartelt, Margret Keuper CVPR Workshop on Explainable AI for Computer Vision (XAI4CV), Jun. 2025 Automated image understanding in the context of climate change Katharina Prasse, Isaac Bravo, Stefanie Walter, Margret Keuper Panel on Climate Visual Analysis, Computational Visual Media Research (CVMR), Aug. 2025, Zurich, Switzerland Detecting manipulated visuals: A computational approach in the climate change discourse Isaac Bravo, Katharina Prasse, Stefanie Walter, Margret Keuper Parallel Panel II: Climate and Health-related Science Communication, AISCICOMM, Jun. 2024, Zurich, Switzerland * equal contribution |
Invited Talks |
DCBM: Data-Efficient Visual Concept Bottleneck Models
Reading Group of Isabel Valera, Saarland University, Feb. 2026, Saarbrücken, Germany Foundation models: Exploring opportunities and pitfalls Climate Change Visuals Workshop, C3DS, University of Exeter, Jun. 2024, Exeter, United Kingdom |
Teaching |
Image Processing (CS647) tutor
Thesis supervision for Bachelor and Master students |
Community Service |
Co-Guest Editor. Special Issue on Computational Climate Visual Research in ViMe.
Reviewer at ML Conferences e.g. NeurIPS, CVPR, ECCV, CCAI. ML Reading Group Organizer. Data & Web Science Group, University of Mannheim. GirlsDay Session Organizer. Hands-on session designed for children aged 12–17. |
|
Template by Jon Barron. Leaf icons created by Pixel perfect - Flaticon. Photo credit: Joseline Weinberg. |